A new synthetic approach toward bacterial transglycosylase substrates, lipid II and lipid IV

Article abstract of DOI:10.1021/ol201806d

A new synthetic approach toward the bacterial transglycosylase substrates, Lipid II (1) and Lipid IV (2), is described. The key disaccharide was synthesized using the concept of relative reactivity value (RRV) and elaborated to Lipid II and Lipid IV by co

Full text of DOI:10.1021/ol201806d

ORGANIC
LETTERS
2011
Vol. 13, No. 17
4600–4603
A New Synthetic Approach toward
Bacterial Transglycosylase Substrates,
Lipid II and Lipid IV
Hao-Wei Shih,†,‡ Kuo-Ting Chen,‡ Ting-Jen R Cheng,‡ Chi-Huey Wong,*,†,‡ and
Wei-Chieh Cheng*,‡
The Genomics Research Center, Academia Sinica, Nankang, Taipei, Taiwan, and
Department of Chemistry, National Taiwan University, Taipei, Taiwan
chwong@gate.sinica.edu.tw; wcheng@gate.sinica.edu.tw
Received July 6, 2011
ABSTRACT
A new synthetic approach toward the bacterial transglycosylase substrates, Lipid II (1) and Lipid IV (2), is described. The key disaccharide was
synthesized using the concept of relative reactivity value (RRV) and elaborated to Lipid II and Lipid IV by conjugation with the appropriate
oligopeptides and pyrophosphate lipids. Interestingly, the results from our HPLC-based functional TGase assay suggested Lipid IV has a higher
affinity for the enzyme than Lipid II.
The increasing prevalence of drug-resistant infections has
stimulated the pharmaceutical industry and scientific com-
munity to identify novel drug targets and develop new
antibiotics against such resistant strains of bacteria
as vancomycin-resistant Enterococci (VRE), methicillin-
resistant Staphylococcus aureus (MRSA), and even the newly
found bacterium with the NDM-1 resistant gene.1 In bac-
terial cell wall biosynthesis, the enzyme transglycosylases
† National Taiwan University.
‡ The Genomics Research Center, Academia Sinica.
(1) (a) Fischbach, M. A.; Walsh, C. T. Science 2009, 325, 1089–1093.
(b) Marshall, E. Science 2008, 321, 364–364. (c) Cheng, T.-J. R.; Wu,
Y.-T.; Yang, S.-T.; Lo, K.-H.; Chen, S.-K.; Chen, Y.-H.; Huang, W.-I.;
Yuan, C.-H.; Guo, C.-W.; Huang, L.-Y.; Chen, K.-T.; Shih, H.-W.;
Cheng, Y.-S. E.; Cheng, W.-C.; Wong, C.-H. Bioorg. Med. Chem. 2010,
Figure 1. The bacterial transglycosylase (TGase) catalyzes its
18, 8512–8529. (d) Yong, D.; Toleman, M. A.; Giske, C. G.; Cho, H. S.;
substrates, Lipid II, Lipid IV, and others, to form peptidoglycan.
Sundman, K.; Lee, K.; Walsh, T. R. Antimicrob. Agents Chemother.
2009, 53, 5046–5054.
(2) (a) Strynadka, N. C. J.; Lovering, A. L.; de Castro, L. H.; Lim, D.
Science 2007, 315, 1402–1405. (b) Halliday, J.; McKeveney, D.; Muldoon,
C.; Rajaratnam, P.; Meutermans, W. Biochem. Pharmacol. 2006, 71, 957–
967. (c) Silver, L. L. Curr. Opin. Microbiol. 2003, 6, 431–438. (d) Goffin,
C.; Ghuysen, J. M. Microbiol. Mol. Biol. Rev. 1998, 62, 1079–1093.
(3) (a) de Kruijff, B.; van Dam, V.; Breukink, E. Prostaglandins,
Leukotrienes Essent. Fatty Acids 2008, 79, 117–121. (b) Cheng, T.-J. R.;
Sung, M.-T.; Liao, H.-Y.; Chang, Y.-F.; Chen, C.-W.; Huang, C.-Y.;
Chou, L.-Y.; Wu, Y.-D.; Chen, Y.; Cheng, Y.-S. E.; Wong, C.-H.; Ma, C.;
Cheng, W.-C. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 431–436. (c) Terrak,
M.; Sauvage, E.; Derouaux, A.; Dehareng, D.; Bouhss, A.; Breukink, E.;
Jeanjean, S.; Nguyen-Disteche, M. J. Biol. Chem. 2008, 283, 28464–28470.
(TGases) use Lipid II (1) to elongate the sugar backbone
with another substrate such as Lipid II, Lipid IV (2), or a
longer TGase substrate and construct the polymerized
peptidoglycan chain (Figure 1).2 Importantly, these bacter-
ial TGases have similar binding motifs and conserved
catalytic residues in most Gram-positive and Gram-negative
bacteria, including drug-resistant bacteria, even though the
peptide moieties of the peptidoglycan chain have mutated.3
r
10.1021/ol201806d
Published on Web 07/28/2011
2011 American Chemical Society

Accordingly, TGase is considered a potential drug target for
antibiotic discovery and development.2,3
peptides, and to modulate the reactivity of the pendant
thioglycosides via an inductive effect. To avoid an annoying
aglycone-transfer side reaction, which occurred by the mis-
match of the relative energies of thioglycosides,7 six possible
monothioglycoside-based building blocks 3À8 were pre-
pared based on our design and existing RRV database.6
After quantitative determination of their RRVs
(Figure 2), we discovered that a TBS group on the C4-
oxygen of 5 dramatically enhanced the glycosyl reactivity
(RRV = 134.1) and that thioglycosides 3 and 4 had much
lower RRVs. The building block 8 (RRV = 9), bearing an
electron-withdrawing chloroacetyl group (ClAc) on O3, had
a lower RRV compared with the lactylated 6 (18.9) and the
unprotected 7 (21.6). Through our evaluation, building
blocks 5and 8were chosen for our oligosaccharide synthesis.
Treatment of 5 and 8 with NIS-TMSOTf gave disacchar-
ide 9 in almost quantitative yield (>90%) (Scheme 1). The
β-linkage for this newly formed glycosidic bond was con-
Unfortunately, the study of TGase is hampered by an
acute shortage of substrates.3 Isolation from natural
sources is limited by the low natural abundance,3 and the
assembly of their complex structures is a very challenging
task for synthetic chemists. Currently, a few studies related
to the synthesis of Lipid II and its analogues,4 and only one
for Lipid IV, have been reported.5 Novel synthetic strate-
gies toward TGase substrates therefore are of great interest.
Our laboratory has already described and exemplified
the utility of the RRV (relative reactivity value) concept for
the synthesis of natural oligosaccharides and antigens.6
Herein, we describe a new synthetic method to prepare
Lipid II and Lipid IV via key oligosaccharides, the design
of which is based on our RRV concept.
Our retrosynthetic analysis is illustrated in Figure 2.
Importantly the correct choice of protecting groups or
substituents for the key disaccharide would first be investi-
gated. For example, the group at the C3 position in the first
sugar plays two pivotal roles: to further conjugate with
1
firmed by the H NMR spectrum (δ 5.17, J = 8.4 Hz).
Scheme 1. Synthesis of Disaccharide 9 and Tetrasaccharide 12
Disaccharide 9was an intermediate in our Lipid II synthesis,
while the two requisite disaccharides 10 and 11 were inter-
mediates for our Lipid IV synthesis. Deprotection of 9either
with NaOMe in MeOH (to give 10) or with TBAF (to give 11)
(4) (a) Schwartz, B.; Markwalder, J. A.; Wang, Y. J. Am. Chem. Soc.
2001, 123, 11638–11643. (b) VanNieuwenhze, M. S.; Mauldin, S. C.; Zia-
Ebrahimi, M.; Winger, B. E.; Hornback, W. J.; Saha, S. L.; Aikins, J. A.;
Blaszczak, L. C. J. Am. Chem. Soc. 2002, 124, 3656–3660. (c) Ye, X. Y.;
Lo, M. C.; Brunner, L.; Walker, D.; Kahne, D.; Walker, S. J. Am. Chem.
Soc. 2001, 123, 3155–3156. (d) Hasper, H. E.; de Kruijff, B.; Breukink, E.
Biochemistry 2004, 43, 11567–11575. (e) Liu, C.-Y.; Guo, C.-W.; Chang,
Y.-F.; Wang, J.-T.; Shih, H.-W.; Hsu, Y.-F.; Chen, C.-W.; Chen, S.-K.;
Wang, Y.-C.; Cheng, T.-J. R.; Ma, C.; Wong, C.-H.; Fang, J.-M.;
Cheng, W.-C. Org. Lett. 2010, 12, 1608–1611.
(5) Zhang, Y.; Fechter, E. J.; Wang, T. S. A.; Barrett, D.; Walker, S.;
Kahne, D. E. J. Am. Chem. Soc. 2007, 129, 3080–3081.
(6) (a) Zhang, Z. Y.; Ollmann, I. R.; Ye, X. S.; Wischnat, R.; Baasov,
T.; Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734–753. (b) Mong,
K. K. T.; Wong, C. H. Angew. Chem., Int. Ed. 2002, 41, 4087–4090. (c)
Ritter, T. K.; Mong, K. K. T.; Liu, H. T.; Nakatani, T.; Wong, C. H.
Angew. Chem., Int. Ed. 2003, 42, 4657–4660.
(7) (a) Li, Z. T.; Gildersleeve, J. C. J. Am. Chem. Soc. 2006, 128,
11612–11619. (b) Li, Z. T.; Gildersleeve, J. C. Tetrahedron Lett. 2007, 48,
559–562.
Figure 2. Our synthetic strategy for Lipid II and Lipid IV. The
RRVs of new desired monothioglycosides were determined as
previously described.6
Org. Lett., Vol. 13, No. 17, 2011
4601

took place in very good yield (Scehem 1). As expected, 10,
without the ClAc group (RRV = 31.3), was found to be
more reactive than 9(4.2) and11 (5.3). Asufficientreactivity
difference between 10 and 11 allowed glycosylation to
proceed to deliver tetrasaccharide 12 in good yield (71%),
without any side products being observed.8 The configura-
tion of the newly fomed glycosidic bond in 12 was identified
as the β-anomer by NMR, as described above.
Scheme 2. Preparation of Lipid II (1)
With the key oligosaccharides 9 and 12 in hand, atten-
tion turned to the synthesis of the target TGase substrates
Lipid II (1) and Lipid IV (2). Unfortunately, initial
attempts to deprotect the N-phthalimide moiety of 9 under
several conditions (Table 1) such as NH2NH2 (entry 1)
and base-mediated conditions (entries 2 and 3) were
unsuccessful.4,9 A three-step, one-pot procedure (NaBH4,
AcOH, and Ac2O) was eventually developed, in which
the N-phthaloyl group of 9 was directly converted to the
requisite N-acetyl group via reduction, intramolecular
lactonization, and acetylation, with concomitant removal
of the ClAc moiety (entries 5 and 6).10 It is notable that use
of a mixed solvent system (THF/i-BuOH/H2O) gave the
best results (86%, Table 1, entry 6).
Table 1. Conditions of N-Phthalimide Deprotection
entry
conditions
yield (%)d
1
2
NH2NH2 H2O (20 equiv), EtOH, reflux, 48 h
6 N KOH, EtOH, reflux, 24 h
À
À
phosphoroimidazolidate (C55PIm) was coupled to 17 in the
presence of 1H-tetrazole. After global deprotection of the
peptidyl protecting groups, Lipid II (1) wasobtainedin37%
yield.
3a 6 N KOH, EtOH, μ.w., 1 h
trace
3
4b (NH2CH2)2 (20 equiv), t-BuOH, 90 °C, 24 h
5
NaBH4 (10 equiv), MeOH, rt, 24 h; AcOH,
71
90 °C, 3 h; then Ac2O, rt, 0.5 h
NaBH4 (10 equiv), THF/i-BuOH/H2O (2:6:1), rt,
24 h; AcOH, 90 °C, 3 h; then Ac2O, rt, 0.5 h
6c
86
Similarly, Lipid IV (2) was synthesized from 12 in an over-
all yield of approximately 5%. A new Lipid IV based fluore-
scent probe 19 was also prepared by reacting the ε-NH2
of the lysine side chain with NBD-X-OSu (succinimidyl
6-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)hexanoate)
fluorophore and applied to our TGase study (Scheme 3).4,11
In an HPLC-based functional enzyme assay,11 E. coli
PBP1b TGase was used as the enzyme model and the
reaction was monitored by the consumption of 19 (see
details in the Supporting Information).
When the assay contained 19 alone, no transglycosyla-
tion occurred (Figure 3). In contrast, when another sub-
strate Lipid II (1) was added, 19 was consumed. These
observations suggest the disaccharide building block pro-
vided by Lipid II is required for chain elongation in E. coli
PBP1b TGase and are consistent with Kahne’s report,
using a radiolabeled Lipid IV.5 Surprisingly, 19 (1 equiv)
was completely consumed even in the presence of an excess
(1.5 equiv) of 1 (Figure 3e), suggesting that Lipid IV may
a The reaction was performed in a sealed tube at 120 W. b TBS group
was also removed. c 20 equiv of NaBH4 were used for 12. d Isolated yield.
As shown in Scheme 2, treatment of 13 with methyl (S)-
lactate triflate under basic conditions gave 14 in 72% yield.
Thiocresol deprotection (NIS, acetone/H2O) of 14, fol-
lowed by phosphorylation,4 provided phosphoryldiester
15 as a single diastereomer. The R-configuration of 15
was confirmed by NMR spectroscopy (3JH1,H2 = 3.2 Hz).
After successive fluoride-mediated desilylation, hydrolysis,
conjugation with tetrapeptide 16, and debenzylation, the
pentapeptidyl phosphodisaccharide 17 was generated
in 42% overall yield from 15. Finally, the undecaprenyl
(8) Grayson, E. J.; Ward, S. J.; Hall, A. L.; Rendle, P. M.; Gamblin,
D. P.; Batsanov, A. S.; Davis, B. G. J. Org. Chem. 2005, 70, 9740–9754.
(9) (a) Debenham, J. S.; Rodebaugh, R.; FraserReid, B. J. Org.
Chem. 1997, 62, 4591–4600. (b) Lidstrom, P.; Tierney, J.; Wathey, B.;
Westman, J. Tetrahedron 2001, 57, 9225–9283.
(11) (a) Shih, H.-W.; Chen, K.-T.; Chen, S.-K.; Huang, C.-Y.;
Cheng, T.-J. R.; Ma, C.; Wong, C.-H.; Cheng, W.-C. Org. Biomol.
Chem. 2010, 8, 2586–2593. (b) Schwartz, B.; Markwalder, J. A.; Seitz,
S. P.; Wang, Y.; Stein, R. L. Biochemistry 2002, 41, 12552–12561.
(10) Osby, J. O.; Martin, M. G.; Ganem, B. Tetrahedron Lett. 1984,
25, 2093–2096.
4602
Org. Lett., Vol. 13, No. 17, 2011

Scheme 3. Preparation of Lipid IV (2) and Its Fluorescent
Analogue (19)
Figure 3. Measurement of the transglycosylation reaction by
HPLC. E. coli PBP1b was incubated with 19 (1 equiv) and
various amounts of 1 (aÀe: 0, 0.3, 0.6, 0.9, 1.5 equiv) in 2 h and
then stopped by adding Moenomycin A (100 μM). In trace f,
Moenomycin A (100 μM) was added first to stop the reaction
(see details in the Supporting Information).
to the final targets in a practical and convergent manner.
Lipid IV was conjugated with a fluorotag to generate a
versatile fluorescent TGase probe. Preliminary biological
studies suggest that Lipid IV may bind to E. coli PBP1b
with a higher affinity than Lipid II. Syntheses of longer
substrates or substrate-based inhibitors using this ap-
proach are currently in progress.
bind to the TGase enzyme with a higher affinity than Lipid II.
It was also found that the rate of consumption of 19
showed a positive dependence with the amount of 1. In
Figure 3, all reactions were terminated by a well-known
TGase inhibitor, Moenomycin A.3,12
In conclusion, an RRV-based thioglycosylation strategy
has been successfully used to design and prepare the key dis-
accharide and tetrasaccharide intermediates of Lipid II and
Lipid IV. These intermediates were subsequently elaborated
Acknowledgment. We thank the National Science
Council and Academia Sinica for financial support.
Supporting Information Available. Experimental proce-
dures, characterization of the synthetic compounds, and
HPLC analysis of transglycosylation. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
(12) In this biological system with probe 19 as a substrate, Moeno-
mycin A exhibited an IC50 value of 23 nM, and the detailed experiment
was shown in the Supporting Information.
Org. Lett., Vol. 13, No. 17, 2011
4603